Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    publication years

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    SplashScreen
    0.92
     Doremi
    0.91
    휋
    0.87
    匁
    0.87
     SplashScreen
    0.85
    Girl
    0.81
    Encryption
    0.81
     Ranbir
    0.79
    Virtual
    0.79
     каттоосуз
    0.79
    POSITIVE LOGITS
     abuse
    0.74
     defending
    0.73
     normalerweise
    0.69
    चक
    0.65
     defends
    0.64
     book
    0.64
     đáp
    0.63
     attacking
    0.63
     rev
    0.62
     defended
    0.61
    Act Density 0.020%

    No Known Activations